Files
catonline_ai/vw-document-ai-indexer/config.yaml
2025-09-26 17:15:54 +08:00

28 lines
925 B
YAML

- data_path: "blob sas url"
datasource_name: "demo-vw-03"
data_dir: ""
base_path: "D:\\tmp\\"
process_file_num: 0
process_file_last_modify: "2025-06-24 00:00:00"
chunk_size: 2048
token_overlap: 128
index_schemas:
- index_name: "index-dev-figure-01-chunk"
data_type: ["metadata", "document", "chunk"]
field_type: "append"
upload_batch_size: 50
fields: ["filepath", "title"]
full_metadata_vector_fields: ["full_headers", "doc_metadata"]
semantic_config_name: "default"
vector_config_name: "vectorSearchProfile"
update_by_field: "filepath"
vector_fields:
- field: "contentVector"
append_fields: ["content"]
- field: "full_metadata_vector"
append_fields: ["full_headers", "doc_metadata"]
merge_fields:
- key: "doc_metadata"
fields: ["title"]
full_metadata_vector_fields: ["full_headers", "doc_metadata"]